The technical paper titled “Low-Cost Ensembling for Deep Neural Network-Based Non-Intrusive Load Monitoring (NILM)” based on the work of the undergraduate project team, B. Gowrienanthan, K. D. I. S. Ratnayake, and N. Kiruthihan, jointly supervised by Dr. V. Logeeshan and Prof. Sisil Kumarawadu in the Department of Electrical Engineering has won the Best Paper Award in the category of Neural Networks at the IEEE World Artificial Intelligence and Internet of Things Congress held in Seatle, USA, 6-9 June, 2022.
NILM is the process of monitoring the power consumption of individual electrical appliances by disaggregating the aggregate power consumption data from a single sensor placed at the point of common coupling leading to better energy savings and demand-side management. The low-cost method proposed in the paper for ensembling deep neural network models trained for the task of load disaggregation does not require the training of multiple different models.
